Pith. sign in

REVIEW 4 major objections 4 minor 37 references

Direct Imaging of Temperature Evolution of Polar Nanoregions and Chemically Ordered Regions in PMN Relaxor: Evidence for Polar Phase Percolation

T0 review · 4 major / 4 minor · reviewed 2026-08-02 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Nanoscale maps show polar clusters in the relaxor PMN grow and interconnect on cooling, pointing to a ferroelectric nanodomain ground state rather than a frozen dipole glass.

desk verdict Direct real-space PNR/COR maps are a step forward; the percolation claim needs more than two temperatures and a 20-nm projection. read the letter →

arxiv 2607.13364 v1 pith:SBGZS2PR submitted 2026-07-15 cond-mat.mtrl-sci

Kohei Hino , Daisuke Morikawa , Desheng Fu , Mitsuru Itoh , Kenji Tsuda This is my paper · ORCID
classification cond-mat.mtrl-sci PACS 77.80.-e77.84.-s68.37.Ma
keywords polarnanoregionschemicallyorderedregionsrelaxorferroelectricsPMNpercolationtransitionconvergent-beamelectrondiffraction4D-STEMrandom-fieldmodel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper reports direct real-space imaging of polar nanoregions (PNRs) and chemically ordered regions (CORs) in the relaxor ferroelectric Pb(Mg1/3Nb2/3)O3 using convergent-beam electron diffraction combined with 4D scanning transmission electron microscopy. At room temperature, PNRs are isolated 2–5 nm clusters, but at 100 K they grow to roughly 10 nm and merge into interconnected networks, while CORs stay static at 2–5 nm and appear to suppress PNR growth. The authors interpret this percolative growth as direct evidence for the random-field model and as incompatible with the dipole-glass scenario, concluding that PMN undergoes a temperature-driven transition to a frustrated ferroelectric nanodomain state near 220 K. If correct, this resolves a long-standing debate about the nature of the low-temperature phase in prototypical relaxors.

What carries the argument

The central technique is convergent-beam electron diffraction (CBED) performed in a 4D-STEM raster scan, which records a full diffraction pattern at every probe position. PNRs are identified by a symmetry-breaking index that measures deviation of the CBED pattern from two-fold rotational symmetry, while CORs are visualized by the intensity of superlattice reflections arising from B-site cation ordering. Because the non-polar matrix, PNRs, and CORs each have distinct symmetry signatures, the method separates their spatial distributions at nanometer resolution and traces how they change with temperature.

What would settle it

Take a thickness series of the same PMN specimen at 100 K: if the interconnected PNR network breaks up into isolated islands when the foil is thinned below roughly 10 nm, the percolation claim would be a projection artifact. Additionally, acquire CBED maps from a reliably non-polar, centrosymmetric crystal and compute the same symmetry-breaking index; if its noise floor exceeds the color-scale minimum used to define PNRs (about 30%), the PNR map cannot be trusted.

Watch

Extended reading notes

Core claim

The paper claims that in PMN, chemically ordered regions with Fm3-m symmetry remain fixed in size and spatial distribution between 295 K and 100 K, while polar nanoregions with R3m symmetry evolve from isolated ~5 nm objects at room temperature into interconnected structures ~10 nm in size at 100 K. Because the PNR and COR maps come from the same 4D-STEM dataset and the same sample area, the authors can directly correlate the two: PNRs appear in regions away from CORs, and CORs act as pinning centers that hinder but do not prevent PNR growth. The observed merging of PNRs into a connected network is presented as decisive against the dipole-glass picture of randomly frozen, non-interacting dip

Load-bearing premise

The maps assume that convergent-beam patterns recorded through a ~20 nm thick specimen predominantly reflect the net local symmetry even when multiple PNRs and CORs overlap along the beam path, and the symmetry-breaking index used to locate PNRs has no stated noise floor or classification threshold.

Editorial extensions

If this is right

  • If the percolation claim holds, PMN's low-temperature state is a frustrated ferroelectric nanodomain state, consistent with the recovery of an underdamped ferroelectric soft mode below ~220 K.
  • The static, pinning role of CORs implies that chemical ordering, not just random B-site valence disorder, controls the scale and connectivity of polar order in relaxors.
  • Direct visualization of PNR growth provides a real-space benchmark for diffuse-scattering models; the irregular, three-dimensional nanodomains observed here differ from the 'pancake-shaped' PNR morphology inferred from earlier scattering studies.
  • The simultaneous mapping of PNRs and CORs in the same specimen region shows a spatial anti-correlation, supporting the view that CORs inhibit local polarization formation rather than nucleate it.
  • The distinction between dipole-glass and random-field behavior has practical implications for designing relaxor-based capacitors and piezoelectrics: the ground state is not a frozen glass but a tunable ferroelectric network.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the measurements were made on a ~20 nm thick foil, the apparent percolation at 100 K could partly be a projection effect: if several PNRs overlap along the beam direction, their symmetry-breaking signals may superimpose and mimic connectivity even if the regions are isolated in three dimensions.
  • The symmetry-breaking index lacks a stated noise floor or classification threshold, so a systematic control measurement on a known non-polar, centrosymmetric crystal region would test how reliably the index distinguishes weak PNR signals from experimental noise.
  • The same 4D-STEM-CBED approach could be extended to other relaxors, such as PZN or PMN-PT, to check whether percolative PNR growth and COR pinning are generic features or specific to PMN.
  • If CORs are true pinning centers, then spatially patterning the COR distribution in a relaxor could provide a route to engineering the percolation temperature and the functional dielectric response.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The manuscript reports 4D-STEM-CBED maps of polar nanoregions (PNRs) and chemically ordered regions (CORs) in the relaxor PMN at 295 K and 100 K. PNRs are identified by a symmetry-breaking index quantifying deviation from two-fold rotational symmetry in CBED patterns (Eq. 1), and CORs by superlattice reflection intensity. The authors claim that CORs remain static with temperature, while PNRs grow from isolated 2–5 nm objects at 295 K to interconnected ~10 nm structures at 100 K, indicating a percolation transition near Tf ≈ 220 K. They further assert that this observation supports the random-field model and 'definitively rule[s] out the dipole glass scenario.'

Significance. If the central claim holds, the paper would provide a direct real-space visualization of PNR evolution and support for a percolation-driven ferroelectric nanodomain ground state in a canonical relaxor, resolving a long-standing debate. The combination of 4D-STEM and CBED to map both polar and chemical order in the same specimen region is methodologically promising, and the representative experimental CBED patterns are convincing evidence for local symmetry breaking. However, the strength of the conclusions considerably exceeds the evidence: the dataset consists of only two temperatures, the analysis rests on a projection-thickness assumption that is not quantitatively validated, and the classification thresholds are not defined. If the results are reproducible with additional temperatures, controlled noise analysis, and overlap simulations, the significance would be high; as presented, the claims outrun the data.

major comments (4)
  1. [Methods (specimen thickness) and Eq. (1)] The central claim of PNR growth and percolation at 100 K depends on interpreting the symmetry-breaking index map as a lateral map of PNRs. But CBED integrates through the entire ~20 nm specimen thickness, and the authors acknowledge that multiple PNRs/CORs may overlap along the beam. In a 20 nm slab with 2–5 nm PNRs, several layers will be projected. Two overlapping PNRs with opposite polarizations can cancel the symmetry-breaking signal, while aligned PNRs will appear as a single larger object, creating apparent connectivity. The statement that CBED 'predominantly reflects the net local symmetry' is asserted, not demonstrated. To support the percolation interpretation, the authors should present multi-slice CBED simulations of realistic 3D arrangements of PNRs (including random, non-percolating configurations) and show that the projected symmetry-breaking maps at 295 K would not produce
  2. [Experimental (temperature conditions) and Fig. 4] Only two temperatures are measured: 295 K and 100 K. The abstract, introduction, and conclusion assert a 'temperature evolution' and a percolation transition 'near 220 K,' but no data exist across or below Tf. From two endpoints one cannot establish a transition, a percolation threshold, or the statement that PNRs 'evolve from isolated...to interconnected.' The strong conclusion that this 'definitively rules out the dipole glass scenario' is not supported by the temporal resolution. At minimum, the authors should add intermediate temperatures (e.g., 220, 180, 150 K) to demonstrate a continuous growth/connectivity transition, or substantially soften the claims to a two-temperature comparison.
  3. [Fig. 2(j) and Eq. (1) — classification thresholds and noise floor] The PNR map is generated from a symmetry-breaking index, but no noise floor, background threshold, or classification criterion is given. The color scale in Fig. 2(j) spans 30%–60%, but the text does not state why 30% corresponds to PNR presence or what value is characteristic of the non-polar matrix. Without a control experiment (e.g., on a known non-polar area) or a noise analysis of the detector and pattern integration, the assignment of bright regions to PNRs is not reproducible. The same applies to the COR map (Fig. 3(g)): the superlattice intensity threshold is not stated. This is a load-bearing issue because the percolation claim rests on the binary classification of pixels as PNR or NPM.
  4. [Fig. 2(c)-(f), (h)-(i) — model dependence of PNR identification] The PNR patterns are simulated using an R3m model with B-site displacements parameterized from prior structural studies [27]. While the breaking of two-fold symmetry is a model-independent observable, the assignment of specific <111> polarization directions to the experimental patterns relies on those simulations. If alternative displacement directions or magnitudes were present, the CBED patterns would change. The paper should explicitly state that the 'R3m with <111> displacements' label is an interpretation based on the assumed model, not a direct measurement, and discuss the sensitivity of the direction assignment to the displacement magnitude. This may not invalidate the maps, but it is a correctness-risk concern for the polarization-direction arrows in Fig. 2(j).
minor comments (4)
  1. [Abstract/Introduction] Typographical and grammatical errors, e.g., 'remains' should be 'remain' in the first paragraph; 'minute' is used ambiguously. The phrase 'The other results obtained by the different specimen area' is awkward and should be rephrased.
  2. [Fig. 2 and Fig. 3 captions] The captions do not specify the exact conditions (e.g., exposure time, probe size) for the experimental patterns in (g)-(i), nor the criteria for selecting the representative probe positions. Adding scale bars to the maps and marking the probe positions clearly would improve reproducibility.
  3. [Supplementary material] The paper references Fig. S1 and two additional sample areas in the Supplementary Material, but the supplementary content is not described in enough detail in the text. For example, it is not stated whether the additional areas show the same percolation behavior.
  4. [Data availability] The data availability statement says data are 'available from the corresponding author upon reasonable request,' but modern 4D-STEM datasets are large; the authors should consider depositing the reduced maps or a representative subset in a public repository to facilitate independent verification.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the experimental imaging chain is self-contained, and the R3m labeling is an independent interpretive model rather than a constructed prediction.

full rationale

I walked the paper's derivation chain: experimental CBED patterns are acquired in 4D-STEM, a symmetry-breaking index is computed from the 180°-rotational comparison of each pattern, PNR maps are built from that index, COR maps from superlattice intensities, and temperature evolution is read directly from the maps. No predicted quantity is fitted from the same data it is later said to predict. The R3m/<111> interpretation is parameterized from prior independent structural studies ([27], not by the present authors), and it is used to label polarization directions, but the central claims of PNR connectivity at 100 K, static CORs, and PNR–COR anti-correlation rest on the directly measured two-fold symmetry breaking and superlattice reflections, not on the R3m parameter values. The self-citations ([22]–[26], [28]) concern the CBED/4D-STEM method and MBFIT software; the paper states the index definition explicitly, and the method is externally established. The projection/overlap caveat about a ~20 nm specimen is an experimental-validity concern (possible stereological artifact), not a circular reduction: the maps are not derived from the percolation conclusion. No equation reduces to another by construction, and no fitted parameter is renamed as a prediction. Therefore no circular step is identified.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical entities. The free parameters are mostly unspecified classification thresholds and a borrowed simulation parameter; the central claims rest on the domain assumptions that CBED symmetry breaking is a clean PNR signature, that thickness-integrated patterns reflect a dominant local phase, and that the two measured temperatures are directly comparable.

free parameters (4)
  • Symmetry-breaking index classification threshold for PNRs = Not stated (color scale 30–60%)
    The PNR maps in Figs. 2(j) and 4(a,d) classify 'bright regions' as PNRs; without an explicit threshold or noise floor, the segmentation is qualitative and operator-dependent.
  • Specimen thickness = ~20 nm
    Estimated by comparing experimental and simulated CBED patterns; used to optimize detection of superlattice reflections and symmetry breaking, and it is load-bearing for the 'net local symmetry' overlap assumption.
  • B-site displacement magnitude in PNR CBED simulations = Taken from prior structural studies [27], value not quoted
    Used to simulate CBED patterns for assigning polarization directions; the exact magnitude is borrowed from prior literature rather than measured in this study.
  • COR superlattice intensity threshold = Not stated (color scale 7.2e6–2.5e7 counts)
    COR maps in Fig. 3(g) and 4(b,e) rely on superlattice intensities; the cutoff distinguishing COR from NPM is not specified.
assumptions (6)
  • domain assumption Dynamical Bloch-wave diffraction theory, as implemented in MBFIT, correctly simulates CBED patterns for ~20 nm PMN at 100 kV.
    Used to classify experimental patterns as NPM, PNR, or COR; errors in the dynamical simulation would propagate directly into the phase assignment.
  • domain assumption In a ~20 nm foil, overlapping PNRs/CORs along the beam direction still produce CBED patterns that 'predominantly reflect the net local symmetry'.
    Stated in the experimental section: this justifies interpreting each probe position as a single dominant local phase, and is load-bearing for the spatial anti-correlation and percolation claims.
  • domain assumption Two-fold rotational symmetry breaking in CBED is a unique signature of polar B-site <111> displacements (R3m), not of other symmetry-lowering effects such as strain, surface relaxation, or defects.
    This underpins the use of the symmetry-breaking index as a PNR map; the paper does not test alternative sources of symmetry lowering.
  • domain assumption PNR displacement parameters from prior structural studies of related relaxors [27] are transferable to PMN.
    Used for simulated CBED patterns to determine polarization directions; the existence of broken symmetry is robust to exact displacement values, but the inferred <111> directions depend on this borrowed model.
  • domain assumption The charge-neutral COR model is representative; the charge-imbalanced model would not change the superlattice detection.
    The paper notes both models yield comparable superlattice intensities (Fig. S1); adopted charge-neutral model does not affect central claim but is an interpretive assumption.
  • domain assumption Differences between the 295 K and 100 K maps are intrinsic to PNR evolution, not caused by specimen drift, beam damage, or changing diffraction conditions.
    The paper acknowledges minor drift and COR intensity variations but attributes them to PNR-induced Bragg-condition changes; this post hoc attribution is load-bearing for the claim that CORs are static.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Direct Imaging of Temperature Evolution of Polar Nanoregions and Chemically Ordered Regions in PMN Relaxor: Evidence for Polar Phase Percolation." pith.science (2026). https://pith.science/paper/SBGZS2PR

@misc{pith2026260713364,
  author       = {Pith},
  title        = {Pith review of: Direct Imaging of Temperature Evolution of Polar Nanoregions and Chemically Ordered Regions in PMN Relaxor: Evidence for Polar Phase Percolation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SBGZS2PR}},
  note         = {Machine review of arXiv:2607.13364}
}
read the original abstract

Polar nanoregions (PNRs) are central to understanding the exceptional dielectric and piezoelectric properties of relaxor ferroelectrics and are key to advancing dielectrics for high-energy storage. However, direct real-space imaging of their formation and evolution remains a major challenge in condensed matter physics. Here, we report the real-space mappings of both PNRs and chemically ordered regions (CORs) in the prototypical relaxor Pb(Mg1/3Nb2/3)O3 and their temperature dependence using convergent-beam electron diffraction (CBED) combined with four-dimensional scanning transmission electron microscopy (4D-STEM). The results reveal that CORs, with sizes of 2-5 nm, remain static with temperature and act to suppress PNR growth. In contrast, PNRs evolve from isolated 2-5 nm regions at room temperature to interconnected structures ~10 nm in size at low temperatures, indicative of a percolation transition. These observations support the random-field model, in which PNRs emerge from a paraelectric matrix and their growth and collective interactions are constrained by random local fields associated with CORs.

Figures

Figures reproduced from arXiv: 2607.13364 by the authors.

Figure 1
Figure 1. FIG 1. Schematic illustrations of local structures in PMN: (a) non-polar matrix (NPM), (b) polar [PITH_FULL_IMAGE:figures/full_fig_p013_1.png] view at source ↗
Figure 2
Figure 2. FIG 2. [110] [PITH_FULL_IMAGE:figures/full_fig_p014_2.png] view at source ↗
Figure 3
Figure 3. FIG 3. [110] [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: FIG 4. (a) Real- [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 2 canonical work pages

  1. [27]

    Dkhil, J

    B. Dkhil, J. M. Kiat, G. Calvarin, G. Baldinozzi, S. B. Vakhrushev, and E. Suard, Phys. Rev. B 65, 024104 (2001)

    2001

  2. [1]

    Burns and F

    G. Burns and F. H. Dacol, Solid State Commun. 42, 9 (1982)

    1982

  3. [2]

    L. E. Cross, Ferroelectrics 151, 305 (1994)

    1994

  4. [3]

    S. E. Park and T. R. Shrout, J. Appl. Phys. 82, 1804 (1997)

    1997

  5. [4]

    S. J. Zhang and F. Li, J. Appl. Phys. 111, 031301 (2012)

    2012

  6. [5]

    A. D. Hilton, D. J. Barber, C. A. Randall, and T. R. Shrout, J. Mater. Sci. 25, 3461 (1990)

    1990

  7. [6]

    Boulesteix, F

    C. Boulesteix, F. Varnier, A. Llebaria, and E. Husson, J. Solid State Chem. 108, 141 (1994)

    1994

  8. [7]

    Husson, M

    E. Husson, M. Chubb, and A. Morell, Mater. Res. Bull. 23, 357 (1988)

    1988

Show all 37 references
  1. [8]

    I. K. Jeong, T. W. Darling, J. K. Lee, T. Proffen, R. H. Heffner, J. S. Park, K. S. Hong, W. Dmowski, and T. Egami, Phys. Rev. Lett. 94, 147602 (2005)

    2005

  2. [9]

    D. Fu, H. Taniguchi, M. Itoh, and S. Mori, Advances in Ferroelectrics. InTech; 2012. Available from: http://dx.doi.org/10.5772/52139

    2012 doi

  3. [10]

    G. Xu, G. Shirane, J. R. D. Copley, and P. M. Gehring, Phys. Rev. B 69, 064112 (2004)

    2004

  4. [11]

    D. Fu, H. Taniguchi, M. Itoh, S. Y. Koshihara, N. Yamamoto, and S. Mori, Phys. Rev. Lett. 103, 207601 (2009)

    2009

  5. [12]

    E. V. Colla, E. Y. Koroleva, N. M. Okuneva, and S. B. Vakhrushev, Phys. Rev. Lett. 74, 1681 (1995)

    1995

  6. [13]

    Levstik, Z

    A. Levstik, Z. Kutnjak, C. Filipic, and R. Pirc, Phys. Rev. B 57, 11204 (1998)

    1998

  7. [14]

    Pirc and R

    R. Pirc and R. Blinc, Phys. Rev. B 60, 13470 (1999)

    1999

  8. [15]

    Bobnar, Z

    V. Bobnar, Z. Kutnjak, R. Pirc, R. Blinc, and A. Levstik, Phys. Rev. Lett. 84, 5892 (2000)

    2000

  9. [16]

    V. V. Westphal, W. Kleemann, and M. D. Glinchuk, Phys. Rev. Lett. 68, 847 (1992)

    1992

  10. [17]

    Fisch, Phys

    R. Fisch, Phys. Rev. B 67, 094110 (2003)

    2003

  11. [18]

    Moriya, H

    Y. Moriya, H. Kawaji, T. Tojo, and T. Atake, Phys. Rev. Lett. 90, 205901 (2003)

    2003

  12. [19]

    B. P. Burton, E. Cockayne, S. Tinte, and U. V. Waghmare, Phase Transitions 79, 91(2006)

    2006

  13. [20]

    Pasciak, M

    M. Pasciak, M. Wolcyrz, and A. Pietraszko, Phys. Rev. B 76, 014117 (2007)

    2007

  14. [21]

    Kumar, J

    A. Kumar, J. N. Baker, P. C. Bowes, M. J. Cabral, S. Zhang, E. C. Dickey, D. L. Irving, and J. M. LeBeau, Nat. Mater. 20, 62 (2021)

    2021

  15. [22]

    Tsuda, R

    K. Tsuda, R. Sano, and M. Tanaka, Phys. Rev. B 86, 214106 (2012)

    2012

  16. [23]

    Tsuda, A

    K. Tsuda, A. Yasuhara, and M. Tanaka, Appl. Phys. Lett. 103, 082908 (2013)

    2013

  17. [24]

    Tsuda and M

    K. Tsuda and M. Tanaka, Appl. Phys. Express 9, 071501 (2016)

    2016

  18. [25]

    Morikawa, Y

    D. Morikawa, Y. Noguchi, and K. Tsuda, Jpn. J. Appl. Phys. 62, SM1003 (2023)

    2023

  19. [26]

    Tsuda and M

    K. Tsuda and M. Tanaka, Acta Crystallogr. A 55, 939 (1999)

    1999

  20. [28]

    Morikawa, M

    D. Morikawa, M. Ageishi, K. Sato, K. Tsuda, and M. Terauchi, Microscopy (Oxf.) 70, 394 (2021)

    2021

  21. [29]

    Egami, W

    T. Egami, W. Dmowski, S. Teslic, P. K. Davies, I. W. Chen & H. Chen, Ferroelectrics 206, 231(1998)

    1998

  22. [30]

    Akbas and P

    M. Akbas and P. Davies, J. Am. Ceram. Soc. 83, 119 (2000)

    2000

  23. [31]

    J. Chen, H. M. Chan, and M. Harmer,J. Am. Ceram. Soc. 72, 593(1989)

    1989

  24. [32]

    Yan and S

    Y. Yan and S. J. Pennycook, Z. Xu and D. Viehland, Appl. Phys. Lett. 72, 3145 (1998)

    1998

  25. [33]

    M. J. Cabral, S. Zhang, E. C. Dickey, and J. M. LeBeau, Appl. Phys. Lett. 112, 082901 (2018)

    2018

  26. [34]

    Wakimoto, C

    S. Wakimoto, C. Stock, R. J. Birgeneau, Z. G. Ye, W. Chen, W. J. L. Buyers, P. M. Gehring, and G. Shirane, Phys. Rev. B 65, 172105 (2002)

    2002

  27. [35]

    G. Xu, Z. Zhong, H. Hiraka, and G. Shirane, Phys. Rev. B 70, 174109(2004)

    2004

  28. [36]

    I.-W. Chen, P. Li, and Y. Wang, J. Phys. Chem Solids 57, 1525 (1996)

    1996

  29. [37]

    I.-W. Chen, J. Phys. Chem Solids 61, 197-208 (2000). This is the author’s peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset. PLEASE CITE THIS ARTICLE AS DOI: 10.1063/5.0297426

    2000 doi

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.